aboutsummaryrefslogtreecommitdiffstatshomepage
path: root/arch/s390/mm/vmem.c
diff options
context:
space:
mode:
Diffstat (limited to 'arch/s390/mm/vmem.c')
-rw-r--r--arch/s390/mm/vmem.c259
1 files changed, 174 insertions, 85 deletions
diff --git a/arch/s390/mm/vmem.c b/arch/s390/mm/vmem.c
index c2583f921ca8..85cddf904cb2 100644
--- a/arch/s390/mm/vmem.c
+++ b/arch/s390/mm/vmem.c
@@ -11,8 +11,11 @@
#include <linux/list.h>
#include <linux/hugetlb.h>
#include <linux/slab.h>
+#include <linux/sort.h>
+#include <asm/page-states.h>
#include <asm/cacheflush.h>
#include <asm/nospec-branch.h>
+#include <asm/ctlreg.h>
#include <asm/pgalloc.h>
#include <asm/setup.h>
#include <asm/tlbflush.h>
@@ -30,11 +33,15 @@ static void __ref *vmem_alloc_pages(unsigned int order)
return memblock_alloc(size, size);
}
-static void vmem_free_pages(unsigned long addr, int order)
+static void vmem_free_pages(unsigned long addr, int order, struct vmem_altmap *altmap)
{
+ if (altmap) {
+ vmem_altmap_free(altmap, 1 << order);
+ return;
+ }
/* We don't expect boot memory to be removed ever. */
if (!slab_is_available() ||
- WARN_ON_ONCE(PageReserved(virt_to_page(addr))))
+ WARN_ON_ONCE(PageReserved(virt_to_page((void *)addr))))
return;
free_pages(addr, order);
}
@@ -44,8 +51,10 @@ void *vmem_crst_alloc(unsigned long val)
unsigned long *table;
table = vmem_alloc_pages(CRST_ALLOC_ORDER);
- if (table)
- crst_table_init(table, val);
+ if (!table)
+ return NULL;
+ crst_table_init(table, val);
+ __arch_set_page_dat(table, 1UL << CRST_ALLOC_ORDER);
return table;
}
@@ -61,6 +70,7 @@ pte_t __ref *vmem_pte_alloc(void)
if (!pte)
return NULL;
memset64((u64 *)pte, _PAGE_INVALID, PTRS_PER_PTE);
+ __arch_set_page_dat(pte, 1);
return pte;
}
@@ -150,7 +160,8 @@ static bool vmemmap_unuse_sub_pmd(unsigned long start, unsigned long end)
/* __ref: we'll only call vmemmap_alloc_block() via vmemmap_populate() */
static int __ref modify_pte_table(pmd_t *pmd, unsigned long addr,
- unsigned long end, bool add, bool direct)
+ unsigned long end, bool add, bool direct,
+ struct vmem_altmap *altmap)
{
unsigned long prot, pages = 0;
int ret = -ENOMEM;
@@ -166,11 +177,11 @@ static int __ref modify_pte_table(pmd_t *pmd, unsigned long addr,
if (pte_none(*pte))
continue;
if (!direct)
- vmem_free_pages((unsigned long) pfn_to_virt(pte_pfn(*pte)), 0);
+ vmem_free_pages((unsigned long)pfn_to_virt(pte_pfn(*pte)), get_order(PAGE_SIZE), altmap);
pte_clear(&init_mm, addr, pte);
} else if (pte_none(*pte)) {
if (!direct) {
- void *new_page = vmemmap_alloc_block(PAGE_SIZE, NUMA_NO_NODE);
+ void *new_page = vmemmap_alloc_block_buf(PAGE_SIZE, NUMA_NO_NODE, altmap);
if (!new_page)
goto out;
@@ -207,7 +218,8 @@ static void try_free_pte_table(pmd_t *pmd, unsigned long start)
/* __ref: we'll only call vmemmap_alloc_block() via vmemmap_populate() */
static int __ref modify_pmd_table(pud_t *pud, unsigned long addr,
- unsigned long end, bool add, bool direct)
+ unsigned long end, bool add, bool direct,
+ struct vmem_altmap *altmap)
{
unsigned long next, prot, pages = 0;
int ret = -ENOMEM;
@@ -224,15 +236,15 @@ static int __ref modify_pmd_table(pud_t *pud, unsigned long addr,
if (!add) {
if (pmd_none(*pmd))
continue;
- if (pmd_large(*pmd)) {
+ if (pmd_leaf(*pmd)) {
if (IS_ALIGNED(addr, PMD_SIZE) &&
IS_ALIGNED(next, PMD_SIZE)) {
if (!direct)
- vmem_free_pages(pmd_deref(*pmd), get_order(PMD_SIZE));
+ vmem_free_pages(pmd_deref(*pmd), get_order(PMD_SIZE), altmap);
pmd_clear(pmd);
pages++;
} else if (!direct && vmemmap_unuse_sub_pmd(addr, next)) {
- vmem_free_pages(pmd_deref(*pmd), get_order(PMD_SIZE));
+ vmem_free_pages(pmd_deref(*pmd), get_order(PMD_SIZE), altmap);
pmd_clear(pmd);
}
continue;
@@ -240,7 +252,7 @@ static int __ref modify_pmd_table(pud_t *pud, unsigned long addr,
} else if (pmd_none(*pmd)) {
if (IS_ALIGNED(addr, PMD_SIZE) &&
IS_ALIGNED(next, PMD_SIZE) &&
- MACHINE_HAS_EDAT1 && addr && direct &&
+ MACHINE_HAS_EDAT1 && direct &&
!debug_pagealloc_enabled()) {
set_pmd(pmd, __pmd(__pa(addr) | prot));
pages++;
@@ -255,7 +267,7 @@ static int __ref modify_pmd_table(pud_t *pud, unsigned long addr,
* page tables since vmemmap_populate gets
* called for each section separately.
*/
- new_page = vmemmap_alloc_block(PMD_SIZE, NUMA_NO_NODE);
+ new_page = vmemmap_alloc_block_buf(PMD_SIZE, NUMA_NO_NODE, altmap);
if (new_page) {
set_pmd(pmd, __pmd(__pa(new_page) | prot));
if (!IS_ALIGNED(addr, PMD_SIZE) ||
@@ -269,12 +281,12 @@ static int __ref modify_pmd_table(pud_t *pud, unsigned long addr,
if (!pte)
goto out;
pmd_populate(&init_mm, pmd, pte);
- } else if (pmd_large(*pmd)) {
+ } else if (pmd_leaf(*pmd)) {
if (!direct)
vmemmap_use_sub_pmd(addr, next);
continue;
}
- ret = modify_pte_table(pmd, addr, next, add, direct);
+ ret = modify_pte_table(pmd, addr, next, add, direct, altmap);
if (ret)
goto out;
if (!add)
@@ -289,27 +301,19 @@ out:
static void try_free_pmd_table(pud_t *pud, unsigned long start)
{
- const unsigned long end = start + PUD_SIZE;
pmd_t *pmd;
int i;
- /* Don't mess with any tables not fully in 1:1 mapping & vmemmap area */
- if (end > VMALLOC_START)
- return;
-#ifdef CONFIG_KASAN
- if (start < KASAN_SHADOW_END && KASAN_SHADOW_START > end)
- return;
-#endif
pmd = pmd_offset(pud, start);
for (i = 0; i < PTRS_PER_PMD; i++, pmd++)
if (!pmd_none(*pmd))
return;
- vmem_free_pages(pud_deref(*pud), CRST_ALLOC_ORDER);
+ vmem_free_pages(pud_deref(*pud), CRST_ALLOC_ORDER, NULL);
pud_clear(pud);
}
static int modify_pud_table(p4d_t *p4d, unsigned long addr, unsigned long end,
- bool add, bool direct)
+ bool add, bool direct, struct vmem_altmap *altmap)
{
unsigned long next, prot, pages = 0;
int ret = -ENOMEM;
@@ -325,7 +329,7 @@ static int modify_pud_table(p4d_t *p4d, unsigned long addr, unsigned long end,
if (!add) {
if (pud_none(*pud))
continue;
- if (pud_large(*pud)) {
+ if (pud_leaf(*pud)) {
if (IS_ALIGNED(addr, PUD_SIZE) &&
IS_ALIGNED(next, PUD_SIZE)) {
pud_clear(pud);
@@ -336,7 +340,7 @@ static int modify_pud_table(p4d_t *p4d, unsigned long addr, unsigned long end,
} else if (pud_none(*pud)) {
if (IS_ALIGNED(addr, PUD_SIZE) &&
IS_ALIGNED(next, PUD_SIZE) &&
- MACHINE_HAS_EDAT2 && addr && direct &&
+ MACHINE_HAS_EDAT2 && direct &&
!debug_pagealloc_enabled()) {
set_pud(pud, __pud(__pa(addr) | prot));
pages++;
@@ -346,10 +350,10 @@ static int modify_pud_table(p4d_t *p4d, unsigned long addr, unsigned long end,
if (!pmd)
goto out;
pud_populate(&init_mm, pud, pmd);
- } else if (pud_large(*pud)) {
+ } else if (pud_leaf(*pud)) {
continue;
}
- ret = modify_pmd_table(pud, addr, next, add, direct);
+ ret = modify_pmd_table(pud, addr, next, add, direct, altmap);
if (ret)
goto out;
if (!add)
@@ -364,29 +368,20 @@ out:
static void try_free_pud_table(p4d_t *p4d, unsigned long start)
{
- const unsigned long end = start + P4D_SIZE;
pud_t *pud;
int i;
- /* Don't mess with any tables not fully in 1:1 mapping & vmemmap area */
- if (end > VMALLOC_START)
- return;
-#ifdef CONFIG_KASAN
- if (start < KASAN_SHADOW_END && KASAN_SHADOW_START > end)
- return;
-#endif
-
pud = pud_offset(p4d, start);
for (i = 0; i < PTRS_PER_PUD; i++, pud++) {
if (!pud_none(*pud))
return;
}
- vmem_free_pages(p4d_deref(*p4d), CRST_ALLOC_ORDER);
+ vmem_free_pages(p4d_deref(*p4d), CRST_ALLOC_ORDER, NULL);
p4d_clear(p4d);
}
static int modify_p4d_table(pgd_t *pgd, unsigned long addr, unsigned long end,
- bool add, bool direct)
+ bool add, bool direct, struct vmem_altmap *altmap)
{
unsigned long next;
int ret = -ENOMEM;
@@ -405,7 +400,7 @@ static int modify_p4d_table(pgd_t *pgd, unsigned long addr, unsigned long end,
goto out;
p4d_populate(&init_mm, p4d, pud);
}
- ret = modify_pud_table(p4d, addr, next, add, direct);
+ ret = modify_pud_table(p4d, addr, next, add, direct, altmap);
if (ret)
goto out;
if (!add)
@@ -418,29 +413,20 @@ out:
static void try_free_p4d_table(pgd_t *pgd, unsigned long start)
{
- const unsigned long end = start + PGDIR_SIZE;
p4d_t *p4d;
int i;
- /* Don't mess with any tables not fully in 1:1 mapping & vmemmap area */
- if (end > VMALLOC_START)
- return;
-#ifdef CONFIG_KASAN
- if (start < KASAN_SHADOW_END && KASAN_SHADOW_START > end)
- return;
-#endif
-
p4d = p4d_offset(pgd, start);
for (i = 0; i < PTRS_PER_P4D; i++, p4d++) {
if (!p4d_none(*p4d))
return;
}
- vmem_free_pages(pgd_deref(*pgd), CRST_ALLOC_ORDER);
+ vmem_free_pages(pgd_deref(*pgd), CRST_ALLOC_ORDER, NULL);
pgd_clear(pgd);
}
static int modify_pagetable(unsigned long start, unsigned long end, bool add,
- bool direct)
+ bool direct, struct vmem_altmap *altmap)
{
unsigned long addr, next;
int ret = -ENOMEM;
@@ -449,6 +435,9 @@ static int modify_pagetable(unsigned long start, unsigned long end, bool add,
if (WARN_ON_ONCE(!PAGE_ALIGNED(start | end)))
return -EINVAL;
+ /* Don't mess with any tables not fully in 1:1 mapping & vmemmap area */
+ if (WARN_ON_ONCE(end > VMALLOC_START))
+ return -EINVAL;
for (addr = start; addr < end; addr = next) {
next = pgd_addr_end(addr, end);
pgd = pgd_offset_k(addr);
@@ -462,7 +451,7 @@ static int modify_pagetable(unsigned long start, unsigned long end, bool add,
goto out;
pgd_populate(&init_mm, pgd, p4d);
}
- ret = modify_p4d_table(pgd, addr, next, add, direct);
+ ret = modify_p4d_table(pgd, addr, next, add, direct, altmap);
if (ret)
goto out;
if (!add)
@@ -475,14 +464,16 @@ out:
return ret;
}
-static int add_pagetable(unsigned long start, unsigned long end, bool direct)
+static int add_pagetable(unsigned long start, unsigned long end, bool direct,
+ struct vmem_altmap *altmap)
{
- return modify_pagetable(start, end, true, direct);
+ return modify_pagetable(start, end, true, direct, altmap);
}
-static int remove_pagetable(unsigned long start, unsigned long end, bool direct)
+static int remove_pagetable(unsigned long start, unsigned long end, bool direct,
+ struct vmem_altmap *altmap)
{
- return modify_pagetable(start, end, false, direct);
+ return modify_pagetable(start, end, false, direct, altmap);
}
/*
@@ -490,7 +481,8 @@ static int remove_pagetable(unsigned long start, unsigned long end, bool direct)
*/
static int vmem_add_range(unsigned long start, unsigned long size)
{
- return add_pagetable(start, start + size, true);
+ start = (unsigned long)__va(start);
+ return add_pagetable(start, start + size, true, NULL);
}
/*
@@ -498,7 +490,8 @@ static int vmem_add_range(unsigned long start, unsigned long size)
*/
static void vmem_remove_range(unsigned long start, unsigned long size)
{
- remove_pagetable(start, start + size, true);
+ start = (unsigned long)__va(start);
+ remove_pagetable(start, start + size, true, NULL);
}
/*
@@ -511,21 +504,25 @@ int __meminit vmemmap_populate(unsigned long start, unsigned long end, int node,
mutex_lock(&vmem_mutex);
/* We don't care about the node, just use NUMA_NO_NODE on allocations */
- ret = add_pagetable(start, end, false);
+ ret = add_pagetable(start, end, false, altmap);
if (ret)
- remove_pagetable(start, end, false);
+ remove_pagetable(start, end, false, altmap);
mutex_unlock(&vmem_mutex);
return ret;
}
+#ifdef CONFIG_MEMORY_HOTPLUG
+
void vmemmap_free(unsigned long start, unsigned long end,
struct vmem_altmap *altmap)
{
mutex_lock(&vmem_mutex);
- remove_pagetable(start, end, false);
+ remove_pagetable(start, end, false, altmap);
mutex_unlock(&vmem_mutex);
}
+#endif
+
void vmem_remove_mapping(unsigned long start, unsigned long size)
{
mutex_lock(&vmem_mutex);
@@ -538,7 +535,7 @@ struct range arch_get_mappable_range(void)
struct range mhp_range;
mhp_range.start = 0;
- mhp_range.end = VMEM_MAX_PHYS - 1;
+ mhp_range.end = max_mappable - 1;
return mhp_range;
}
@@ -561,33 +558,125 @@ int vmem_add_mapping(unsigned long start, unsigned long size)
}
/*
- * map whole physical memory to virtual memory (identity mapping)
- * we reserve enough space in the vmalloc area for vmemmap to hotplug
- * additional memory segments.
+ * Allocate new or return existing page-table entry, but do not map it
+ * to any physical address. If missing, allocate segment- and region-
+ * table entries along. Meeting a large segment- or region-table entry
+ * while traversing is an error, since the function is expected to be
+ * called against virtual regions reserved for 4KB mappings only.
*/
+pte_t *vmem_get_alloc_pte(unsigned long addr, bool alloc)
+{
+ pte_t *ptep = NULL;
+ pgd_t *pgd;
+ p4d_t *p4d;
+ pud_t *pud;
+ pmd_t *pmd;
+ pte_t *pte;
+
+ pgd = pgd_offset_k(addr);
+ if (pgd_none(*pgd)) {
+ if (!alloc)
+ goto out;
+ p4d = vmem_crst_alloc(_REGION2_ENTRY_EMPTY);
+ if (!p4d)
+ goto out;
+ pgd_populate(&init_mm, pgd, p4d);
+ }
+ p4d = p4d_offset(pgd, addr);
+ if (p4d_none(*p4d)) {
+ if (!alloc)
+ goto out;
+ pud = vmem_crst_alloc(_REGION3_ENTRY_EMPTY);
+ if (!pud)
+ goto out;
+ p4d_populate(&init_mm, p4d, pud);
+ }
+ pud = pud_offset(p4d, addr);
+ if (pud_none(*pud)) {
+ if (!alloc)
+ goto out;
+ pmd = vmem_crst_alloc(_SEGMENT_ENTRY_EMPTY);
+ if (!pmd)
+ goto out;
+ pud_populate(&init_mm, pud, pmd);
+ } else if (WARN_ON_ONCE(pud_leaf(*pud))) {
+ goto out;
+ }
+ pmd = pmd_offset(pud, addr);
+ if (pmd_none(*pmd)) {
+ if (!alloc)
+ goto out;
+ pte = vmem_pte_alloc();
+ if (!pte)
+ goto out;
+ pmd_populate(&init_mm, pmd, pte);
+ } else if (WARN_ON_ONCE(pmd_leaf(*pmd))) {
+ goto out;
+ }
+ ptep = pte_offset_kernel(pmd, addr);
+out:
+ return ptep;
+}
+
+int __vmem_map_4k_page(unsigned long addr, unsigned long phys, pgprot_t prot, bool alloc)
+{
+ pte_t *ptep, pte;
+
+ if (!IS_ALIGNED(addr, PAGE_SIZE))
+ return -EINVAL;
+ ptep = vmem_get_alloc_pte(addr, alloc);
+ if (!ptep)
+ return -ENOMEM;
+ __ptep_ipte(addr, ptep, 0, 0, IPTE_GLOBAL);
+ pte = mk_pte_phys(phys, prot);
+ set_pte(ptep, pte);
+ return 0;
+}
+
+int vmem_map_4k_page(unsigned long addr, unsigned long phys, pgprot_t prot)
+{
+ int rc;
+
+ mutex_lock(&vmem_mutex);
+ rc = __vmem_map_4k_page(addr, phys, prot, true);
+ mutex_unlock(&vmem_mutex);
+ return rc;
+}
+
+void vmem_unmap_4k_page(unsigned long addr)
+{
+ pte_t *ptep;
+
+ mutex_lock(&vmem_mutex);
+ ptep = virt_to_kpte(addr);
+ __ptep_ipte(addr, ptep, 0, 0, IPTE_GLOBAL);
+ pte_clear(&init_mm, addr, ptep);
+ mutex_unlock(&vmem_mutex);
+}
+
void __init vmem_map_init(void)
{
- phys_addr_t base, end;
- u64 i;
-
- for_each_mem_range(i, &base, &end)
- vmem_add_range(base, end - base);
- __set_memory((unsigned long)_stext,
- (unsigned long)(_etext - _stext) >> PAGE_SHIFT,
- SET_MEMORY_RO | SET_MEMORY_X);
- __set_memory((unsigned long)_etext,
- (unsigned long)(__end_rodata - _etext) >> PAGE_SHIFT,
- SET_MEMORY_RO);
- __set_memory((unsigned long)_sinittext,
- (unsigned long)(_einittext - _sinittext) >> PAGE_SHIFT,
- SET_MEMORY_RO | SET_MEMORY_X);
- __set_memory(__stext_amode31, (__etext_amode31 - __stext_amode31) >> PAGE_SHIFT,
- SET_MEMORY_RO | SET_MEMORY_X);
-
- /* lowcore must be executable for LPSWE */
+ __set_memory_rox(_stext, _etext);
+ __set_memory_ro(_etext, __end_rodata);
+ __set_memory_rox(_sinittext, _einittext);
+ __set_memory_rox(__stext_amode31, __etext_amode31);
+ /*
+ * If the BEAR-enhancement facility is not installed the first
+ * prefix page is used to return to the previous context with
+ * an LPSWE instruction and therefore must be executable.
+ */
if (!static_key_enabled(&cpu_has_bear))
set_memory_x(0, 1);
-
+ if (debug_pagealloc_enabled()) {
+ /*
+ * Use RELOC_HIDE() as long as __va(0) translates to NULL,
+ * since performing pointer arithmetic on a NULL pointer
+ * has undefined behavior and generates compiler warnings.
+ */
+ __set_memory_4k(__va(0), RELOC_HIDE(__va(0), ident_map_size));
+ }
+ if (MACHINE_HAS_NX)
+ system_ctl_set_bit(0, CR0_INSTRUCTION_EXEC_PROTECTION_BIT);
pr_info("Write protected kernel read-only data: %luk\n",
(unsigned long)(__end_rodata - _stext) >> 10);
}
Copyright © 1996 – 2023 Jason A. Donenfeld. All Rights Reverse Engineered.